CN116203965A - AGV-based path planning algorithm for improving ant colony algorithm - Google Patents
AGV-based path planning algorithm for improving ant colony algorithm Download PDFInfo
- Publication number
- CN116203965A CN116203965A CN202310269020.8A CN202310269020A CN116203965A CN 116203965 A CN116203965 A CN 116203965A CN 202310269020 A CN202310269020 A CN 202310269020A CN 116203965 A CN116203965 A CN 116203965A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- path
- agv
- ant colony
- grid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 51
- 239000003016 pheromone Substances 0.000 claims abstract description 21
- 238000004519 manufacturing process Methods 0.000 claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 9
- 230000007547 defect Effects 0.000 claims abstract 2
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 230000004888 barrier function Effects 0.000 description 4
- 238000004088 simulation Methods 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 241000257303 Hymenoptera Species 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a path planning algorithm based on an AGV, and provides a path planning algorithm of an improved ant colony algorithm (Ant Colony Optimization, ACO) aiming at the structure and characteristics of the AGV in a production workshop. Aiming at the defects that the ant colony algorithm has low efficiency, low convergence rate, easy trapping in local optimum and the like, an ASACO algorithm which combines an A-algorithm and the ant colony algorithm is provided, a grid map is firstly established by using a grid method according to the topographic environment of a production workshop, a starting point and an end point are established according to the grid map, then a predicted path from the starting point to the end point is established by using a heuristic A-algorithm, the pheromone concentration of the ant colony algorithm is initialized through the predicted path, the early blind search of the ant colony algorithm is avoided, the turning angle and the turning times of the path are taken as consideration factors aiming at the simple path distance in actual conditions, and finally the running path which is most suitable for the AGV is found through the ant colony algorithm. The comparison test shows that the algorithm has obvious optimizing effect.
Description
Technical Field
The invention relates to a path planning algorithm based on an AGV (automatic guided vehicle) and used for improving an ant colony algorithm, and belongs to the field of artificial intelligence.
Background
With the development of the internet of things technology, an automatic guided vehicle (automated guided vehicle, AGV) is widely applied to warehouse logistics. AGVs refer to transport vehicles equipped with autonomous positioning and navigation devices, the primary function of which is material handling in workshops, which are critical devices for job shop logistics transport, playing a critical role in the process of shop processing and transport.
The large-scale AGV is applied to the industries of the Internet of things, the machinery manufacturing industry, the express logistics industry and the like. Modern logistics scheduling systems have become an integral part of various industries, and large enterprises establish intelligent logistics systems which take AGVs as the dominant ones. Hundreds of AGV trolleys are used for establishing a logistics system without manpower in the whole process of warehousing, loading, sorting and unloading, so that the efficiency and the flexibility of workshops are greatly improved.
In the past, path planning is an important ring in AGV technology, and in the working environment with obstacles or in the unknown field, the shortest directed path from the current position to the target position is searched, so that the aim of realizing the optimization target is fulfilled. The general optimization objective is to meet the requirements of the minimum total time for completing the job, the minimum travel path of the AGV when completing the objective task, the minimum energy consumption of the AGV when completing the task, etc. The optimal path of the AGV is planned by taking the up-down slope and the turning times of the path as consideration factors, not only aiming at the simple path distance.
Disclosure of Invention
In order to achieve the above purpose, the present invention provides a path planning algorithm based on an improved ant colony algorithm of an AGV, comprising the following steps:
(1) Establishing a grid map with barriers through a grid method;
(2) Improving the search direction and fitness function value of the A-algorithm, improving the search efficiency, and changing the pheromone in the grid points according to the estimated path;
(3) Initializing the concentration of the ant colony pheromone according to the estimated path of the A-algorithm, and enhancing the early-stage searching efficiency of the ant colony;
(4) An optimal path is planned for the number of path turns and AGV path collisions.
The step (1) specifically comprises the following steps:
(1.1) scanning a map of the production plant environment, and creating a rasterized pattern according to the obstacle.
(1.2) the environment map is rasterized herein, abstracted into an n-order matrix form, the grids are sequentially numbered for convenience of processing and description of the algorithm, the horizontal axis is named as an X-axis, the vertical axis is named as a Y-axis, and the coordinates corresponding to the corresponding grids are (X i ，y i ) The calculation method of the grid number Z is shown in formula 1.
After processing, each grid has its own sequence number: 1,2，3，4，5，6，7，8，9…n 2
Z＝X i +Y i (i<＝n) (1)
(1.3) As shown in the matrix, a non-passing barrier grid through which the AGV can normally pass is shown by 0, and a non-passing barrier grid is shown by 1.
The step (2) specifically comprises the following steps:
and (2.1) the algorithm A selects the node with the smallest estimated value as the expansion node of the next neighborhood by comparing the estimated values of the expansion nodes. The heuristic function can quickly find a path with the shortest distance in the grid map, and the heuristic function can be represented by the formula (2):
f(n)＝g(n)+h(n) (2)
wherein n is the current grid number; g (n) is the known shortest length for the starting grid to reach the current grid n; h (n) is the distance from the current grid n to the target grid, and is generally expressed by Euclidean distance.
(2.2) the heuristic function Euclidean used therein is expressed as formula (3):
where n.x is the abscissa of node n, n.y is the ordinate of node n, target.x is the abscissa of the target point, and target.y is the ordinate of the target point.
(2.3) in the conventional a-algorithm, adding feasible non-traversed neighboring grids of each current grid into the to-be-traversed grid set, and then traversing the grids one by one according to the value of the heuristic function of the to-be-traversed grid, thereby generating a large number of invalid traversals. According to the method, a weight coefficient is added in the aspect of the pre-estimated function to promote the search of the pre-estimated direction in the early stage, so that invalid traversal is avoided. It can be expressed by formula (4):
f(n)＝g(n)+w×h(n) (4)
the w coefficient depends on the size of the heuristic function h, and the larger the value of w is when h is, the higher the importance of the heuristic function is when the heuristic function is far away from the target node, and the search direction is helped to be more close to the target node.
And (2.4) obtaining a predicted path according to the improved A-x algorithm, marking grid points on the path, and changing the concentration of pheromones.
The step (3) comprises the following steps:
(3.1) the search node of the ant colony algorithm indicates a transition from the current node to the next node, and the transition probability thereof can be represented by the formula (5):
in the middle of
For the transition probability of transitioning from the current node i to the next node j, τ ij (t) is the pheromone concentration between two nodes, eta ij (t) is a heuristic function whose value is the inverse of the distance between two nodes, where the distance is also denoted by Euclidean distance.
(3.2) after each ant completes a path, updating the pheromone of the node, wherein the pheromone updating formula can be represented by a formula (6):
τ in ij (t+1) represents the pheromone concentration after the pheromone update; τ ij (t) represents the concentration of the pheromone before the pheromone update; ρ is the pheromone volatility coefficient.
(3.3) initializing pheromones for all the rasterized nodes to be C, marking the node set thereof as R according to the pre-estimated path marked before, and changing the initializing pheromones thereof, wherein the initializing pheromones can be shown by a formula (7):
where e is a positive number.
The step (4) comprises the following steps:
the method for searching the path by utilizing the ant colony algorithm considers that the turning times in the path can generate larger abrasion to the AGV, so the turning times are reduced as much as possible while planning, and meanwhile, the larger the turning angle of the AGV is, the damage to the use of the AGV is also caused, so more turning times and smaller turning angles are avoided in path planning. Heuristic function improvement to the ant colony algorithm can be shown by formula (8):
wherein the method comprises the steps of
Representing a turning factor heuristic function ++>
As a heuristic function of turning angle, x 1 ,y 1 ,y 1 For the coefficients of each heuristic function,
in the path turning factors, the traveling direction is stored by using a matrix through grid marks shown in fig. 1, and in order to obtain a planned path with smaller turning times, a turning factor heuristic function is introduced as shown in a formula (9):
wherein e is a constant coefficient, sigma is a straight-going priority coefficient, and the value is assigned to the actual running map i Representing all feasible next nodes, move i Representing the current driving direction, move j And the running direction of the next node is indicated, if the two directions are the same, the straight running is continued, and at the moment, the turning factors guide ants to continue straight running, so that the situation that the distance is short to find a path with more turns is avoided. When the AGV has a turning path, the angle of the turning should be selected as small as possible to avoid overlarge turning angle caused by smaller turning times,
drawings
FIG. 1 rasterized matrix diagram
FIG. 2 AGV path planning flowchart
FIG. 3 comparison of iteration curves
FIG. 4 is a graph of iteration count versus time
Detailed Description
The invention will be further described with reference to the drawings.
Step 1: the grid pattern is established according to the space environment where the AGVs are located by using a grid method according to the scene of an actual production workshop as shown in fig. 1, wherein black matrix blocks represent barrier grids which cannot be passed by the AGVs, white matrix blocks quickly represent paths through which the AGVs can pass, and each matrix block can completely contain the AGVs and reserve a proper amount of safety spacing. The rasterized map is then matrix converted with 0 and 1, where 0 represents the path that the AGV can pass through and 1 represents the obstacle grid that the AGV cannot pass through.
Step 2: as shown in fig. 2, the driving direction of the path is determined according to the obstacle distribution in the grid matrix and the driving start point and end point of the AGV, the estimated path conforming to the start point and the end point is found according to the improved a-algorithm, the initialized pheromone concentration of the path is modified, and the path node with optimal number of times of turning and optimal turning angle in the path is found iteratively by using the ASACO algorithm according to the initialized pheromone concentration until the optimal path between the start point and the end point is found.
According to the scale of the grid map, an ACO algorithm and an ASACO algorithm are respectively adopted to carry out path planning, and as shown in fig. 3, green dots in the grid map are used as path starting points and red dots are used as path ending points. The dashed line is the path planning result of the ACO algorithm, the solid line is the path planning result of the ASACO algorithm, and as shown in fig. 3 and fig. 4, compared with the original ant colony algorithm, the Wen Suanfa is obviously improved in the aspects of path optimizing, convergence speed inflection point number and the like. Wherein the simulation experiment data pair is shown in table 1:
table 1 algorithm simulation result data comparison
From comparison of the simulation data in table 1, it is known that the path length obtained by the ASACO algorithm is 27.2596m, the path length is 28.3940m shorter than that of the ACO algorithm, the convergence speed is obviously improved, and the inflection points of the two paths are compared to find that the inflection points of the path obtained by the ASACO algorithm are less.
Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to specific embodiments, and that the embodiments may be combined appropriately to form other embodiments that will be understood by those skilled in the art.
Claims (4)
1. A path planning algorithm in a workshop scene based on an AGV comprises the following specific steps:
(1) The grid map is constructed by the obstacle distribution situation for the actual scene.
(2) And constructing an estimated path by using a heuristic A algorithm according to the starting point and the end point in the map.
(3) The pheromone concentration of the ant colony algorithm is initialized through the estimated path, blind searching in the early stage of the ant colony algorithm is avoided, a model is built by reducing turning times and turning angles as far as possible aiming at the selection of AGVs and paths in actual conditions, and finally the optimal AGV driving path is completed through iteration of the algorithm.
2. The path planning algorithm in an AGV-based production plant scenario of claim 1, wherein: the step (1) specifically comprises the following steps: and scanning an environment map of a production workshop, rasterizing the map according to the distribution condition of the obstacles to construct a rasterized map, abstracting the rasterized map into a matrix form, dividing the sequence number of the rasterized map, and corresponding the position of each grid in a coordinate form.
3. The path planning algorithm in an AGV-based production plant scenario of claim 2, wherein: the step (2) specifically comprises the following steps: according to the grid map matrix obtained in the step 1 and the starting point and the end point corresponding to the AGV, an estimated path from the starting point to the end point can be obtained by utilizing an improved A-based algorithm, and the initialized pheromone concentration is changed by utilizing the estimated path aiming at the defects that the ant colony algorithm searches blindly in the early stage and is easy to fall into a local optimal solution and the like.
4. A path planning algorithm in an AGV-based production plant scenario according to claim 3, wherein: the step (3) specifically comprises the following steps: and (3) carrying out iteration of an ant colony algorithm by utilizing the pheromone concentration obtained after the initialization in the step (2), searching the next grid point by utilizing a roulette and a searching tabu table when searching the considered path each time, and judging the pheromone concentration, the turning times and the turning angles in all feasible grid points to determine the selection of the next grid point because the AGV can damage the use of the AGV in each turning, wherein the method is used for continuously carrying out iteration in the grid points to find the optimal path which is most suitable for the AGV to travel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310269020.8A CN116203965A (en) | 2023-03-19 | 2023-03-19 | AGV-based path planning algorithm for improving ant colony algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310269020.8A CN116203965A (en) | 2023-03-19 | 2023-03-19 | AGV-based path planning algorithm for improving ant colony algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116203965A true CN116203965A (en) | 2023-06-02 |
Family
ID=86507739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310269020.8A Pending CN116203965A (en) | 2023-03-19 | 2023-03-19 | AGV-based path planning algorithm for improving ant colony algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116203965A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116976542A (en) * | 2023-09-22 | 2023-10-31 | 天津万事达物流装备有限公司 | Optimization method and system for automatic logistics sorting path |
| CN117232512A (en) * | 2023-08-04 | 2023-12-15 | 广东工业大学 | Unmanned aerial vehicle path acquisition method for efficient search and nested cooperation optimization strategy |
| CN117670162A (en) * | 2023-12-06 | 2024-03-08 | 珠海市格努信息技术有限公司 | Intelligent logistics solving method in field |
-
2023
- 2023-03-19 CN CN202310269020.8A patent/CN116203965A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117232512A (en) * | 2023-08-04 | 2023-12-15 | 广东工业大学 | Unmanned aerial vehicle path acquisition method for efficient search and nested cooperation optimization strategy |
| CN117232512B (en) * | 2023-08-04 | 2024-05-24 | 广东工业大学 | Unmanned aerial vehicle path acquisition method for efficient search and nested cooperation optimization strategy |
| CN116976542A (en) * | 2023-09-22 | 2023-10-31 | 天津万事达物流装备有限公司 | Optimization method and system for automatic logistics sorting path |
| CN116976542B (en) * | 2023-09-22 | 2023-12-01 | 天津万事达物流装备有限公司 | Optimization method and system for automatic logistics sorting path |
| CN117670162A (en) * | 2023-12-06 | 2024-03-08 | 珠海市格努信息技术有限公司 | Intelligent logistics solving method in field |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116203965A (en) | AGV-based path planning algorithm for improving ant colony algorithm | |
| CN112650229B (en) | Mobile robot path planning method based on improved ant colony algorithm | |
| CN110160546B (en) | Mobile robot path planning method | |
| CN111323016A (en) | Mobile robot path planning method based on self-adaptive ant colony algorithm | |
| CN115079705A (en) | Routing planning method for inspection robot based on improved A star fusion DWA optimization algorithm | |
| CN108444490B (en) | Robot path planning method based on depth fusion of visible view and A-x algorithm | |
| CN112327876B (en) | Robot path planning method based on terminal distance index | |
| CN113867368A (en) | Robot path planning method based on improved gull algorithm | |
| CN113359718B (en) | Method and equipment for fusing global path planning and local path planning of mobile robot | |
| CN115509239B (en) | Unmanned vehicle route planning method based on air-ground information sharing | |
| CN115407784B (en) | Unmanned vehicle route planning method based on air-ground information complementation | |
| CN113110601B (en) | Unmanned aerial vehicle power line inspection path optimization method and device | |
| WO2022142893A1 (en) | Path planning method and apparatus for biped robot, and biped robot | |
| CN115167474A (en) | Mobile robot path planning optimization method | |
| CN114964261A (en) | Mobile robot path planning method based on improved ant colony algorithm | |
| CN110954124A (en) | Adaptive path planning method and system based on A-PSO algorithm | |
| CN115167398A (en) | Unmanned ship path planning method based on improved A star algorithm | |
| CN116540738A (en) | Mobile robot path planning method based on motion constraint improved ant colony algorithm | |
| CN113359721B (en) | Improved A-based AGV path planning method combined with motion control | |
| CN117522078A (en) | Method and system for planning transferable tasks under unmanned system cluster environment coupling | |
| CN116817947B (en) | Random time path planning method based on variable step length mechanism | |
| CN116449846A (en) | Optimization method of ant colony algorithm | |
| Zhao et al. | A compound path planning algorithm for mobile robots | |
| Li et al. | Research on AUV path planning based on improved ant colony algorithm | |
| CN114281105A (en) | Unmanned aerial vehicle path planning method for substation inspection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |